Range-free Selective Multilateration for Anisotropic Wireless Sensor Networks

Shigeng Zhang,Jianxin Wang,Xuan Liu,Jiannong Cao

DOI: https://doi.org/10.1109/secon.2012.6275791

2012-01-01

Abstract:In anisotropic wireless sensor networks, range-free multilateration-based localization (RFML) protocols severely suffer from large error in node-anchor distance estimations or the bad geometry of anchors involved in the localization process. In this paper, we propose selective multilateration (SM), a RFML protocol in which a node adaptively selects a subset of anchor nodes with accurate distance estimates and good geometric distribution to perform multilateration. We make two main contributions: (1) We exploit the locality property of per hop length (two nearby nodes share similar per hop length to a far anchor node) to estimate node-anchor distances. This induces smaller distance estimation error than previous approaches that use a unified average per hop length for all nodes. (2) We propose a method to heuristically select a subset of good anchor nodes that have small distance estimation errors and good geometry quality measured by geometric dilution of precision (GDOP) to perform multilateration. We also use a method which mines sub-hop resolution proximity between neighboring nodes to reduce error in distance estimates which further improves localization accuracy. Simulation results show that, compared with PDM, SM improves distance estimation accuracy by up to 45 percent. The localization accuracy in SM is improved by up to 45 percent and 27 percent in average. Furthermore, SM incurs much less communication and computational overhead than PDM does, making it more suitable for large scale WSNs.